Sell-side research - You get what you pay for, no more, no less

 

Remember, all Sell-Side Research contains at least 1 of the following 3 elements

1.Trades that 40-Act Funds are running after serious traders/HFs stopped out.

2.Death Trap Trades where the bank’s desk needs to take the other side.

3. An honest opinion of an analyst.

Never forget, in life, even lies are intriguing and useful, they reveal where someone's interests are

- Matt Kessel

This view is an extreme, but there is truth in the words that sell-side forecasts and analysis can be biased, are driven by incentives, and have potential conflicts of interest albeit the biases are small as measured by numerous studies. There is also an element of dispersion in forecasts because the quality of analysts varies.

Honest opinions of experts can diverge between being poor and very effective. Academic and private researchers have studied analyst forecasts for decades and we do have a pretty good idea of the benefits and costs of research especially for equity markets. Additionally, there is enough analysis of sell-side forecasts to determine how to weigh this information to generate better forecasts. Don't use single forecasts blindly but employ the wisdom of crowds. Use the crowd estimates as a placeholder for market consensus. Agreement with the crowd may generate positive returns but will not create unique alpha.

In general, research finds that earnings and stock price forecasts may be slightly better than time series forecasts, but they may not be efficient and there may be biases over both under- and over-reaction to different market environments. Analysts have a hard time with turning points and change. Macro forecasts are biased and may not be efficient. There is limited edge gained from using sell-side information.

One of the more current issues with analyzing sell-side research is that the focus has been fairly narrow. There is a difference between reporting or producing analysis and forecasting price or earnings. The analysis comes first and may be more valuable. From analysis, there is an inference or a forecast. If the analysis is poor or assumptions are flawed, it is impossible to get the forecast right. The quality of sell-side research should be centered on the ability of the analyst to provide firm and industry information quickly, cheaply, and efficiently. Providing description of the market details in order to support investor analysis may be more valuable than the work of generating an earnings estimate, stock estimate, or point forecast.